连续、长时间序列的多年冻土温度数据在开展气候变化对多年冻土的影响及其生态、水文效应研究中有着重要的科学意义.本文利用西大滩、五道梁、唐古拉3个观测点的实测地温数据首先对多年冻土模型及其参数化方案进行了验证、优化和标校,以实测数据和经过校正的CMIP6逐月5 cm地表土壤温度数据作为模型的驱动数据,模拟了3个观测点1900～2019年多年冻土地温变化,并对1920年以后的模拟温度变化序列进行分析.结果表明:(1)多年冻土模型对处于地温年变化深度以下地温的模拟误差低于0.1°C,表明模型对于多年冻土的热状态具有较好的模拟能力;(2) 1920～2019年西大滩、五道梁、唐古拉各模拟深度的年平均温度均呈现升温趋势,年地温变化(15 m)处的平均升温速率为0.07°C/10 a(0.05～0.09),不同深度岩土层的热状态对气候变化具有不同的响应时间,深层地温对气候变化的响应相较于浅层有20年左右的滞后;(3)多年冻土上限下降速率相差不大,平均为0.6 cm/a;多年冻土下限的上升速率分别为13.4、4.0和4.0 cm/a;多年冻土厚度分别减少13.9、4.6、4.7 m;(4) 3个观测点...

准确模拟多年冻土地温变化，对深入了解多年冻土土壤温度变化特征、不同下垫面条件的土壤热物性以及对气候变化预测均有重要意义。运用GIPL2模型模拟阿拉斯加不同土壤类型地温，并对其实用性进行了评估。模拟结果表明：在土壤粒径较大的区域，用模型的默认参数可以模拟出土壤温度随季节变化的趋势，且在浅层的模拟值比深层更接近观测值，当深度超过0.5 m时模拟值与观测值存在较大的误差。在土壤质地复杂的区域由于土壤含水率等一系列参数对模型的影响，要准确模拟试验点的温度变化，需要可靠的观测数据。在同一地点安装气象站点和土壤参数传感器来监测多年冻土，可以为评估多年冻土热状况对气候持续变化的响应提供必要的数据。总体而言，当有足够的数据支撑时，GIPL2模型对土壤热状况的模拟精度较好，是一种模拟多年冻土区不同深度土壤热物性较为理想的模型。

The long-term and continuous permafrost temperature data is of great significance to the study of permafrost, climate, ecology, hydrology and engineering on the Qinghai-Tibet Plateau (QTP), but the available observed data sequence is no longer than 25 years. To address the gap, we first attempt to reconstruct the sequences of permafrost temperature at three monitoring sites including Xidatan, Wudaoliang and Tanggula along the QTP engineering corridor from 1920 to 2019, based on one of the most used permafrost models worldwide (i.e., Geophysical Institute Permafrost Lab version 2 (GIPL2)). The GIPL2 model and its parameterized schemes were first evaluated and calibrated using the ground temperature observations at the three sites. The monthly near surface ground temperature at the depth of 5 cm after calibration and correction based on monitoring data was used as forcing dataset to simulate the temperature change of the permafrost from 1900 to 2019. The temperature change sequences since 1920 were selected to discuss the changes of permafrost on the QTP, and its responses to climate change. Results showed that (i) the GIPL2 model can well simulate the thermal state of permafrost on the QTP with low simulation errors (below 0.1 degrees C) at the depth of zero annual amplitude; (ii) the annual average ground temperature at different depths for all three sites experienced warming trends from 1920 to 2019, in which the average warming rate was 0.07 degrees C/10 a (0.05-0.09) at the depth of zero annual amplitude (15 m). Besides, the site with the largest warming rate at the shallow layer (3 m) was found in Wudaoliang, while the deep layer (30 m) was in Xidatan; (iii) the permafrost temperature at the shallow layer increased rapidly since 1980. Nevertheless, the response times of the thermal conditions to climate change varied with soil layers, among which the deep layer lagged by about 20 years compared to the shallow layer; (iv) permafrost thicknesses for the Xidatan, Wudaoliang and Tanggula sites were decreased by 13.9 m, 4.6 m and 4.7 m respectively. The average deepening rate of the permafrost table and rising rates of permafrost base for the three sites were 0.6 cm/a and 10.27 cm/a, respectively. More specifically, the deepening rate of the permafrost table was 0.5 cm/a for Xidatan, 0.6 cm/a for Wudaoliang and 0.7 cm/a for Tanggula, and the rising rate of the permafrost base was 13.4 cm/a for Xidatan and 4.0 cm/a for both Wudaoliang and Tanggula. Compared with that in Wudaoliang and Tanggula, the permafrost in Xidatan was relatively unstable and its response to climate change was more sensitive. Although the simulations of the GIPL2 model could be impacted by the accuracy of the forcing data (e.g., 5 cm ground temperature), the reconstructed permafrost temperature changes from 1920 to 2019 were consistent with the observations over the past 40 years. Besides, our results also confirmed the continuous warming phenomenon of permafrost on the QTP since 1920. These findings can well fill the narrow gap relating to the short sequence and discontinuity of the permafrost temperature dataset on the QTP, and provide a baseline of permafrost changes to the scientific community for a better understanding of the changes in the cryosphere, ecosystem, water resources, and even climate. Nevertheless, some limitations in temperature reconstruction and model processing were noted. In the future, multiple aspects including accurate forcing data and complex factors (e.g., heat convection and lateral heat flow exchange) should be considered comprehensively in the model to reduce the uncertainties of ground temperature simulations.

Borehole-measured soil temperatures have been routinely used to calibrate soil parameters in permafrost modeling, although they are sparse in the Qinghai-Tibet Plateau (QTP). A feasible alternative is to calibrate models using land surface temperatures. However, the quantitative impacts of various soil parameterizations on permafrost modeling remain unexplored. To quantify these impacts, two sets of soil parameters (denoted as Psoil and Psurf) were obtained via calibration using borehole temperature measurement and ERA5-Land (the land component of the fifth generation of European Re-Analysis) skin temperature, respectively, and applied to the Geophysical Institute Permafrost Laboratory Version 2 (GIPL 2.0) model. Comparing against the borehole -measured soil temperatures of 4 soil layers, the ERA5-Psurf simulation (with Root Mean Squared Error, i.e., RMSE from 1.4 degrees C to 3.9 degrees C) outperform ERA5-Psoil simulation (RMSE from 1.4 degrees C to 3.9 degrees C) during 2006-2014. The obtained Psoil and Psurf were then utilized as soil parameters in GIPL 2.0 to model permafrost dynamics for a long period from 1983 to 2019, respectively, using ERA5-Land as forcing data. Simulations revealed significant disparities. In comparison to the simulation using Psurf results using Psoil show that the mean annual soil tem-perature at 1 m depth was 2.72 degrees C lower with a 0.01 degrees C/a (50.0%) lower trend; the active layer thickness was 0.81 m (35.7%) less with a 2.16 cm/a (82.1%) lower trend; the duration of the thawing season at 1 m depth was underestimated by about one month, and the zero-curtain period was about 23 days (37.7%) shorter. The change rates of the zero-curtain period, however, were comparable. This study implies that choosing soil parameteri-zations is critical for model evaluation against observations and long-term model prediction.

Fire is an important factor controlling the composition and thickness of the organic layer in the black spruce forest ecosystems of interior Alaska. Fire that burns the organic layer can trigger dramatic changes in the underlying permafrost, leading to accelerated ground thawing within a relatively short time. In this study, we addressed the following questions. (1) Which factors determine post-fire ground temperature dynamics in lowland and upland black spruce forests? (2) What levels of burn severity will cause irreversible permafrost degradation in these ecosystems? We evaluated these questions in a transient modeling-sensitivity analysis framework to assess the sensitivity of permafrost to climate, burn severity, soil organic layer thickness, and soil moisture content in lowland (with thick organic layers, similar to 80 cm) and upland (with thin organic layers, similar to 30 cm) black spruce ecosystems. The results indicate that climate warming accompanied by fire disturbance could significantly accelerate permafrost degradation. In upland black spruce forest, permafrost could completely degrade in an 18 m soil column within 120 years of a severe fire in an unchanging climate. In contrast, in a lowland black spruce forest, permafrost is more resilient to disturbance and can persist under a combination of moderate burn severity and climate warming.